Optimizing video delivery to multiple users over a wireless channel is a challenging task, especially when the objective of maximizing the spectral efficiency has to be jointly considered with the objective of providing a fair video quality. In this paper a novel cross-layer optimization framework for scalable video delivery over OFDMA wireless systems is proposed, aiming at maximizing the sum of the achievable rates while minimizing the distortion difference among multiple videos. Layering as optimization decomposition is considered to vertically decompose the problem and a novel efficient iterative local approximation (ILA) approach is proposed to evaluate the global solution. The ILA algorithm requires a limited information exchange between the application and the MAC layers, which independently run algorithms that handle parameters and constraints characteristic of a single layer. The numerical evaluations show the fast convergence of the ILA algorithm and demonstrate the significant video quality improvement of the proposed strategy with respect to optimizations based on rate fairness.